In this chapter you will learn about Data Augmentation. You have a set of images, you want to add more images from these existing ones.
How?
Say Image 1 is a single image, you may perform rotation, brightness adjustments, noise addition to it for generating many such images.
How does this help?
Well if the network sees more variations it will learn better. So if you have a vase image with different rotations , brightness and still say its a vase, the network will learn better.
The common techniques:-
a. Rotation
b. Brightness, Contrast
c. Noise
d. Blur
e. Changing background
f. Using GANs
A summary is provided in the GPT link: https://chatgpt.com/share/6a98f2c7-8f04-83e8-8a11-d2f638840b5e
Practice 5: You will find a streamlit based chapter at https://github.com/jhilikb/UCS668
Prerequisite: Make sure that you have the augmentation folder in your course folder UCS668.
Start your python environment (say ucs668) that you set up (in Lab0) for the course using source ucs668/bin/activate (on your terminal if you use Linux/Mac) or using .\ucs668\Scripts\Activate.ps1 (on your powershell if using windows). In both cases you should be in the folder where you created the environment
cd augmentation
streamlit run aug.py
You will see the chapter start on your browser. In your browser, you can now explore the example tasks and understand how to solve them.
Exercise:
a. Explore cv2 blur, noise, convertScaleAbs, rotate functions
b. Formulate how advanced augmentation techniques can be used.